features_extraction.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from models.classifier import model\n",
    "from models.cnn_md import cnn_md"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/aidin/anaconda3/lib/python3.8/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: /home/aidin/anaconda3/lib/python3.8/site-packages/torchvision/image.so: undefined symbol: _ZN2at4_ops19empty_memory_format4callEN3c108ArrayRefINS2_6SymIntEEENS2_8optionalINS2_10ScalarTypeEEENS6_INS2_6LayoutEEENS6_INS2_6DeviceEEENS6_IbEENS6_INS2_12MemoryFormatEEE\n",
      "  warn(f\"Failed to load image Python extension: {e}\")\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "from docopt import docopt\n",
    "from models.classifier import model\n",
    "from preprocessing.dataset import Dataset2Channels, Dataset1Channel\n",
    "from utils.constants import MAPPING_LABELS_DICT, MAPPING_LYING_LABELS\n",
    "from datetime import datetime\n",
    "from torchvision import transforms\n",
    "from utils.torch import ToTensor4D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "TYPE = 'mDoppler'\n",
    "\n",
    "mode = 'standing'\n",
    "aggregate = True\n",
    "train_data = 'train_mDoppler_2channels_standing.h5'\n",
    "\n",
    "test_data ='test_mDoppler_2channels_standing.h5'\n",
    "\n",
    "\n",
    "if mode == 'lying':\n",
    "        labels_transform = np.vectorize(\n",
    "            lambda label: MAPPING_LYING_LABELS[label]\n",
    "        )\n",
    "elif aggregate:\n",
    "    labels_transform = np.vectorize(\n",
    "        lambda label: MAPPING_LABELS_DICT[label]\n",
    "    )\n",
    "else:\n",
    "    labels_transform = None\n",
    "\n",
    "if TYPE == 'mDoppler':\n",
    "    features_transform = transforms.Compose([\n",
    "        #lambda x: x[:, :, 9:-9],\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize((0,), (1,))\n",
    "    ])\n",
    "else:\n",
    "    features_transform = transforms.Compose([\n",
    "        ToTensor4D(),\n",
    "        transforms.Normalize((0,), (1,))\n",
    "    ])\n",
    "\n",
    "train_data = Dataset2Channels(\n",
    "    TYPE=TYPE,\n",
    "    dirname='DATA_preprocessed',\n",
    "    filename=train_data,\n",
    "    features_transform=features_transform,\n",
    "    labels_transform=labels_transform,\n",
    "    combine_channels=True\n",
    ")\n",
    "\n",
    "test_data = Dataset2Channels(\n",
    "    TYPE=TYPE,\n",
    "    dirname='DATA_preprocessed',\n",
    "    filename=test_data,\n",
    "    features_transform=features_transform,\n",
    "    labels_transform=labels_transform,\n",
    "    combine_channels=True\n",
    ")\n",
    "\n",
    "in_channels = 2\n",
    "\n",
    "# shuffle data\n",
    "train_data.shuffle()\n",
    "test_data.shuffle()\n",
    "\n",
    "\n",
    "if labels_transform is not None:\n",
    "    num_classes = len(np.unique(labels_transform(train_data.labels[:])))\n",
    "else:\n",
    "    num_classes = len(np.unique(train_data.labels[:]))\n",
    "\n",
    "# load model\n",
    "model_name = 'CNN-MD'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_features(model, input_data):\n",
    "    # Set the model to evaluation mode\n",
    "    model.eval()\n",
    "\n",
    "    # Get the feature maps from the convolutional layers\n",
    "    feature_maps = []\n",
    "    hooks = []\n",
    "\n",
    "    def hook_fn(module, input, output):\n",
    "        feature_maps.append(output.detach())\n",
    "\n",
    "    # Register the hook to each convolutional layer\n",
    "    for module in model.cnn.modules():\n",
    "        if isinstance(module, torch.nn.Conv2d):\n",
    "            hooks.append(module.register_forward_hook(hook_fn))\n",
    "\n",
    "    # Forward pass to generate the feature maps\n",
    "    model(input_data)\n",
    "\n",
    "    # Unregister the hooks\n",
    "    for hook in hooks:\n",
    "        hook.remove()\n",
    "\n",
    "    # Plot the feature maps\n",
    "    num_layers = len(feature_maps)\n",
    "    fig, axs = plt.subplots(1, num_layers, figsize=(20, 15))\n",
    "\n",
    "    for i, fmap in enumerate(feature_maps):\n",
    "        fmap = fmap.squeeze(0)  # Remove the batch dimension\n",
    "        num_channels = fmap.size(0)\n",
    "        axs[i].imshow(torch.sum(fmap, dim=0), cmap='gray')\n",
    "        axs[i].set_title(f'Layer {i+1}\\nChannels: {num_channels}')\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the model object\n",
    "classifier = model(\n",
    "    train_data=train_data,\n",
    "    test_data=test_data,\n",
    "    case=3,\n",
    "    model_type=model_name,\n",
    "    device='cpu',\n",
    ")\n",
    "\n",
    "dropout = 0.2\n",
    "\n",
    "classifier.create_model(\n",
    "    in_channels=in_channels,\n",
    "    num_classes=num_classes,\n",
    "    dropout=dropout,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "load = 'CNN-MD__0.695__20230611'\n",
    "\n",
    "if load != \"None\":\n",
    "    print(f'Loading model {load}')\n",
    "    classifier.load_model(\n",
    "        name=load,\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size=64\n",
    "classifier.create_DataLoaders(\n",
    "    batch_size=batch_size,\n",
    "    #num_workers=1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifier.summary(\n",
    "    save=True,\n",
    "    name=f'{model_name}_case{3}_summary.txt'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "cnn_md(\n",
       "  (cnn): Sequential(\n",
       "    (0): Conv2d(2, 8, kernel_size=(3, 3), stride=(1, 1), padding=same)\n",
       "    (1): ELU(alpha=1.0)\n",
       "    (2): MaxPool2d(kernel_size=(2, 2), stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "    (3): Dropout2d(p=0.2, inplace=False)\n",
       "    (4): Conv2d(8, 16, kernel_size=(3, 3), stride=(1, 1), padding=same)\n",
       "    (5): ELU(alpha=1.0)\n",
       "    (6): MaxPool2d(kernel_size=(2, 2), stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "    (7): Dropout2d(p=0.2, inplace=False)\n",
       "    (8): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=same)\n",
       "    (9): ELU(alpha=1.0)\n",
       "    (10): MaxPool2d(kernel_size=(2, 2), stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "    (11): Dropout2d(p=0.2, inplace=False)\n",
       "    (12): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=same)\n",
       "    (13): ELU(alpha=1.0)\n",
       "    (14): MaxPool2d(kernel_size=(2, 2), stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "  )\n",
       "  (flatten): Flatten(start_dim=1, end_dim=-1)\n",
       "  (fc): Sequential(\n",
       "    (0): Linear(in_features=1536, out_features=128, bias=True)\n",
       "    (1): ReLU()\n",
       "    (2): Dropout(p=0.2, inplace=False)\n",
       "    (3): Linear(in_features=128, out_features=5, bias=True)\n",
       "    (4): Softmax(dim=1)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier.model.eval()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fd5a3caa700>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(train_data[100][0][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1500 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_features(classifier.model, train_data[100][0].unsqueeze(0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load model\n",
    "model_name = 'CNN-MD'\n",
    "\n",
    "# Create the model object\n",
    "classifier = model(\n",
    "    train_data=train_data,\n",
    "    test_data=test_data,\n",
    "    case=3,\n",
    "    model_type=model_name,\n",
    "    device='cpu',\n",
    ")\n",
    "\n",
    "dropout = 0.2\n",
    "\n",
    "classifier.create_model(\n",
    "    in_channels=in_channels,\n",
    "    num_classes=num_classes,\n",
    "    dropout=dropout,\n",
    ")\n",
    "\n",
    "load = 'ResNet18__0.696__20230610_SAVED'\n",
    "\n",
    "if load != \"None\":\n",
    "    print(f'Loading model {load}')\n",
    "    classifier.load_model(\n",
    "        name=load,\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1500 with 17 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_features(model, input_data):\n",
    "    # Set the model to evaluation mode\n",
    "    model.eval()\n",
    "\n",
    "    # Get the feature maps from the convolutional layers\n",
    "    feature_maps = []\n",
    "    hooks = []\n",
    "\n",
    "    def hook_fn(module, input, output):\n",
    "        feature_maps.append(output.detach())\n",
    "\n",
    "    # Register the hook to each convolutional layer\n",
    "    for module in model.modules():\n",
    "        if isinstance(module, torch.nn.Conv2d):\n",
    "            hooks.append(module.register_forward_hook(hook_fn))\n",
    "\n",
    "    # Forward pass to generate the feature maps\n",
    "    model(input_data)\n",
    "\n",
    "    # Unregister the hooks\n",
    "    for hook in hooks:\n",
    "        hook.remove()\n",
    "\n",
    "    # Plot the feature maps\n",
    "    num_layers = len(feature_maps)\n",
    "    fig, axs = plt.subplots(1, num_layers, figsize=(20, 15))\n",
    "\n",
    "    for i, fmap in enumerate(feature_maps):\n",
    "        fmap = fmap.squeeze(0)  # Remove the batch dimension\n",
    "        num_channels = fmap.size(0)\n",
    "        axs[i].imshow(torch.sum(fmap, dim=0), cmap='gray')\n",
    "        axs[i].set_title(f'Layer {i+1}\\nChannels: {num_channels}')\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "\n",
    "plot_features(classifier.model, train_data[20][0].unsqueeze(0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[[0.2079, 0.1978, 0.1805,  ..., 0.1577, 0.1599, 0.1723],\n",
       "         [0.2081, 0.1973, 0.1788,  ..., 0.1547, 0.1568, 0.1701],\n",
       "         [0.2060, 0.1940, 0.1769,  ..., 0.1555, 0.1516, 0.1626],\n",
       "         ...,\n",
       "         [0.1756, 0.1689, 0.1725,  ..., 0.1601, 0.1757, 0.1929],\n",
       "         [0.1714, 0.1634, 0.1634,  ..., 0.1718, 0.1825, 0.1949],\n",
       "         [0.1688, 0.1585, 0.1556,  ..., 0.1758, 0.1853, 0.1954]],\n",
       "\n",
       "        [[0.1189, 0.1232, 0.1276,  ..., 0.1351, 0.1410, 0.1388],\n",
       "         [0.1160, 0.1215, 0.1262,  ..., 0.1383, 0.1419, 0.1387],\n",
       "         [0.1138, 0.1196, 0.1244,  ..., 0.1423, 0.1413, 0.1399],\n",
       "         ...,\n",
       "         [0.1288, 0.1267, 0.1294,  ..., 0.1308, 0.1378, 0.1433],\n",
       "         [0.1269, 0.1273, 0.1313,  ..., 0.1410, 0.1446, 0.1404],\n",
       "         [0.1269, 0.1278, 0.1313,  ..., 0.1444, 0.1458, 0.1378]]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'ResNet18' object has no attribute 'cnn'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[35], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m plot_features(classifier\u001b[39m.\u001b[39;49mmodel, train_data[\u001b[39m0\u001b[39;49m][\u001b[39m0\u001b[39;49m]\u001b[39m.\u001b[39;49munsqueeze(\u001b[39m0\u001b[39;49m))\n",
      "Cell \u001b[0;32mIn[34], line 16\u001b[0m, in \u001b[0;36mplot_features\u001b[0;34m(model, input_data)\u001b[0m\n\u001b[1;32m     13\u001b[0m     feature_maps\u001b[39m.\u001b[39mappend(output\u001b[39m.\u001b[39mdetach())\n\u001b[1;32m     15\u001b[0m \u001b[39m# Register the hook to each convolutional layer\u001b[39;00m\n\u001b[0;32m---> 16\u001b[0m \u001b[39mfor\u001b[39;00m module \u001b[39min\u001b[39;00m model\u001b[39m.\u001b[39;49mcnn\u001b[39m.\u001b[39mmodules():\n\u001b[1;32m     17\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(module, torch\u001b[39m.\u001b[39mnn\u001b[39m.\u001b[39mConv2d):\n\u001b[1;32m     18\u001b[0m         hooks\u001b[39m.\u001b[39mappend(module\u001b[39m.\u001b[39mregister_forward_hook(hook_fn))\n",
      "File \u001b[0;32m~/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py:1207\u001b[0m, in \u001b[0;36mModule.__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m   1205\u001b[0m     \u001b[39mif\u001b[39;00m name \u001b[39min\u001b[39;00m modules:\n\u001b[1;32m   1206\u001b[0m         \u001b[39mreturn\u001b[39;00m modules[name]\n\u001b[0;32m-> 1207\u001b[0m \u001b[39mraise\u001b[39;00m \u001b[39mAttributeError\u001b[39;00m(\u001b[39m\"\u001b[39m\u001b[39m'\u001b[39m\u001b[39m{}\u001b[39;00m\u001b[39m'\u001b[39m\u001b[39m object has no attribute \u001b[39m\u001b[39m'\u001b[39m\u001b[39m{}\u001b[39;00m\u001b[39m'\u001b[39m\u001b[39m\"\u001b[39m\u001b[39m.\u001b[39mformat(\n\u001b[1;32m   1208\u001b[0m     \u001b[39mtype\u001b[39m(\u001b[39mself\u001b[39m)\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m, name))\n",
      "\u001b[0;31mAttributeError\u001b[0m: 'ResNet18' object has no attribute 'cnn'"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.15"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}